Scholarlink

<p>Healthcare managers continuously strive to enhance service quality and patient satisfaction. Predicting one-year survival post-treatment is crucial for evaluating hospital performance and identifying key survival factors. This study compares nine machine learning models using a dataset of 25,079 patient records from a Greenland hospital with 18 variables. Evaluating multiple models ensures the selection of the best-performing one, a standard practice in machine learning. Notably, XGBoost and Gradient Boosting achieved the highest accuracy, with AUCs of 0.90 and 0.91, respectively. These findings underscore the role of machine learning in improving treatment assessment and guiding data-driven decisions for better patient care.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Predicting the chances of patient’s survival following one year of medical treatment

  • 期刊论文

  • 发表日期：

    2025年6月4日

查看全文
  • Preeti,
  • Neetu Gupta

摘要

Healthcare managers continuously strive to enhance service quality and patient satisfaction. Predicting one-year survival post-treatment is crucial for evaluating hospital performance and identifying key survival factors. This study compares nine machine learning models using a dataset of 25,079 patient records from a Greenland hospital with 18 variables. Evaluating multiple models ensures the selection of the best-performing one, a standard practice in machine learning. Notably, XGBoost and Gradient Boosting achieved the highest accuracy, with AUCs of 0.90 and 0.91, respectively. These findings underscore the role of machine learning in improving treatment assessment and guiding data-driven decisions for better patient care.